rECHOmmend: An ECG-based Machine-learning Approach for Identifying Patients at High-risk of Undiagnosed Structural Heart Disease Detectable by Echocardiography

Circulation, Tempus-authored
Authors Alvaro E. Ulloa-Cerna, Linyuan Jing, John M. Pfeifer, Sushravya Raghunath, Jeffrey A. Ruhl, Daniel B. Rocha, Joseph B. Leader, Noah Zimmerman, Greg Lee, Steven R. Steinhubl, Christopher W. Good, Christopher M. Haggerty, Brandon K. Fornwalt and RuiJun Chen

Background: Timely diagnosis of structural heart disease improves patient outcomes, yet many remain underdiagnosed. While population screening with echocardiography is impractical, electrocardiogram (ECG)-based prediction models can help target high-risk patients. We developed a novel ECG-based machine learning approach to predict multiple structural heart conditions, hypothesizing that a composite model would yield higher prevalence and positive predictive values (PPVs) to facilitate meaningful recommendations for echocardiography.

Methods: Using 2,232,130 ECGs linked to electronic health records and echocardiography reports from 484,765 adults between 1984-2021, we trained machine learning models to predict the presence or absence of any of seven echocardiography-confirmed diseases within one year. This composite label included: moderate or severe valvular disease (aortic/mitral stenosis or regurgitation, tricuspid regurgitation), reduced ejection fraction <50%, or interventricular septal thickness >15mm. We tested various combinations of input features (demographics, labs, structured ECG data, ECG traces) and evaluated model performance using 5-fold cross-validation, multi-site validation trained on one site and tested on 10 independent sites, and simulated retrospective deployment trained on pre-2010 data and deployed in 2010.

Results: Our composite ‘rECHOmmend’ model using age, sex and ECG traces had an area under the receiver operating characteristic curve (AUROC) of 0.91 and PPV of 42% at 90% sensitivity, with a composite label prevalence of 17.9%. Individual disease models had AUROCs from 0.86-0.93 and lower PPVs from 1%-31%. AUROCs for models using different input features ranged from 0.80-0.93, increasing with additional features. Multi-site validation showed similar results to cross-validation, with an aggregate AUROC of 0.91 across our independent test set of 10 clinical sites after training on a separate site. Our simulated retrospective deployment showed that for ECGs acquired in patients without pre-existing structural heart disease in the year 2010, 11% were classified as high-risk, of which 41% (4.5% of total patients) developed true echocardiography-confirmed disease within one year.

Conclusions: An ECG-based machine learning model using a composite endpoint can identify a high-risk population for having undiagnosed, clinically significant structural heart disease while outperforming single disease models and improving practical utility with higher PPVs. This approach can facilitate targeted screening with echocardiography to improve under-diagnosis of structural heart disease.